Vous êtes sur la page 1sur 34


Medulloblastoma in childhood
- revisiting intrathecal therapy in infants and children

Sharon Conroy, Lecturer in Paediatric Clinical Pharmacy
Martin Garnett, Associate Professor Pharmaceutical Sciences
Michael Vloeberghs, Professor Paediatric Neurosurgery
Richard Grundy, Professor Paediatric Neuro-Oncology and Cancer Cell Biology
Ian Craven, Research student
David Walker, Professor Paediatric Oncology

Children's Brain Tumour Research Centre,
University of Nottingham,
University Park,

Corresponding author: Professor David A Walker
Tel: +44 1158 230632.

The authors can confirm that this work is original. It was presented as a poster at the 11

International Symposium on Pediatric Neuro-Oncology and the abstract published as below:

Conroy S, Craven I, Garnett M, Barrett S, Punt J, Parker T, Walker DA. An extensive review of
chemotherapy agents suitable for intrathecal (IT) therapy in primitive neuro ectodermal
tumour (PNET). Abstract EXP 7 from 11
International Symposium on Pediatric Neuro-
Oncology. Neuro-Oncology 2004;6:428

There are no conflicts of interest to declare. This work was conducted without any financial
assistance outside of the University of Nottingham

The manuscript has been read and approved by all the authors. Each author believes that the
manuscript represents honest work.
Medulloblastoma in childhood
- revisiting intrathecal therapy in infants and children

Sharon Conroy, Martin Garnett, Michael Vloeberghs, Richard Grundy, Ian Craven
and David Walker


Introduction: Intrathecal chemotherapy is being explored in medulloblastoma in pre-school
children as part of brain-sparing strategies and as an alternative to unacceptably neurotoxic
cranio-spinal radiotherapy. The range of drugs suitable for this route of administration is
restricted by the lack of research evidence of pharmacological suitability and efficacy of other
drugs in medulloblastoma.

Methods: Ideal clinical, biological, physicochemical and pharmaceutical properties for
intrathecal administration were defined through literature review of pharmaceutical texts,
Medline, Embase and consulting the manufacturers. 126 chemotherapy agents were assessed
against these criteria by searching the academic domain of pharmaceutical texts, computer
databases and consultation with manufacturers.

Results: Of 126 candidates drugs, 99 were rejected because of documentation of their
irritant nature, neurotoxicity and requirement for hepatic activation in standard pharmaceutical
texts. Fifty were rejected for a single identifiable reason including, neurotoxicity (n=24),
irritant (n=15), needs enzyme activation (n=5), clinical evidence of intrathecal neurotoxicity
(n=4) and no evidence of tumour-specific efficacy (n=2). Where two reasons were cited the
justifications were: neurotoxic and irritant (n=3) and needs activation and systemic
administration results in equivalent concentration (n=1). Twenty seven drugs remained of
which 12 were selected as eligible for further clinical investigation, and 15 were selected for
further pre-clinical investigation.

Conclusions: The predetermined criteria were not applicable, in their entirety, in the majority
of drugs, due to lack of information in the academic domain, emphasising the importance of a
more open approach to sharing basic drug information. The prioritised list of 12 candidate
drugs for clinical trial and 15 for pre-clinical investigation justify that a concerted research
effort in this area of practice is made. (317 words)
Medulloblastoma, the commonest malignant CNS tumour of childhood [125], has the capacity
to disseminate through the lepto-meninges, presenting at diagnosis or at relapse after
treatment. Leptomeningeal metastases (LM) grow at the interface of the brain and spinal
fluid. They are thought to occur as a result of interplay between operative disruption of
tumour cells contaminating the CSF and inherent biological capacity for tumour cells to
migrate, adhere and invade neural tissue. Where LM are present at diagnosis biological factors
are increasingly being associated with this dissemination and poorer cure rates with standard
therapy. Systemic chemotherapy has an established role in medulloblastoma with evidence of
chemosensitivity, enhanced survival rates and rising survival rates reported by trials groups
and population registries [80] in recent eras.

The use of intrathecal chemotherapy in childhood leukaemia and CNS tumours has been the
focus of reviews [17, 49, 108] and reports of phase 1 and 2 studies [17]. In leukaemia the
intensification of systemic and intrathecal therapy has permitted exclusion of CNS radiotherapy
for the majority of newly diagnosed cases with very low CNS relapse rates, acceptable late
toxicity using methotrexate alone or in combination with cytosine and hydrocortisone. The
intrathecal route in leukaemia has been demonstrated to bypass the blood brain barrier (BBB),
enhance CSF drug concentrations reducing systemic exposure and, therefore, toxicity.
Selection of drugs suitable for intrathecal administration and with evidence of sensitivity to
primary CNS tumours is a challenge in medulloblastoma, therefore. Preliminary evidence of
the efficacy of intrathecal therapy is emerging. Slavc reported extensive use of intrathecal
therapy at relapse using etoposide, mafosfamide and others [120, 121]. Other studies report
the feasibility of this route of drug administration at relapse and as part of palliative care with
infrequent complications and prolonged symptom free intervals. Rutkowski reported very high
survival rates in pre-school children treated with chemotherapy-only strategy including
intraventricular methotrexate [109]. The proposal that intrathecal therapy be tested in a
prospective European trial of medulloblastoma in this young age group is currently the focus of
debate [139].

In order to further develop an enhanced evidence-base for the selection of drugs that are
effective against medulloblastoma, suitable for administration by the intrathecal route, and
non-toxic to the nervous system, we developed criteria for the ideal intrathecal agent.
Following this a systematic literature review was conducted using these criteria to identify
candidate drugs worthy of further trial for intrathecal use in medulloblastoma.

Identification of candidate drugs
A comprehensive list of chemotherapy agents was extracted from a drug pharmacopoeia [87].
Properties of the ideal intrathecal drug were identified based upon literature review [4, 10, 13,
33, 53, 102, 108, 132, 136, 143, 150, 151] and grouped according to: ideal clinical
properties, ideal biological properties and ideal physico-chemical and pharmaceutical

Ideal Clinical Properties
The drug would need to be non-irritant to avoid chemical arachnoiditis or meningitis. Any drug
reported in any of our reference sources to cause irritation or thrombophlebitis was therefore
eliminated from further consideration.

Any drug given intrathecally has a significant risk of neurotoxicity. Evidence for neurotoxicity
when given systemically or intrathecally was determined as an unacceptable risk for further
evaluation of CSF administration. It was not possible to grade the neurotoxicities as they were
variably reported. Some drugs which are not usually neurotoxic when given systemically at
conventional doses, may be toxic if given intra-arterially, intrathecally or in very high doses
[130]. Any drug reported in any of our reference sources to cause neurotoxicity was therefore
eliminated from further consideration. Methotrexate and cytosine arabinoside, whilst suitable
for IT delivery, have only been studied in medulloblastoma with concurrent systemic
chemotherapy agents, the chemosensitivity of medulloblastoma to these drugs is, therefore,

Tumour sensitivity / mechanism of action
Evidence of activity of the drug against medulloblastoma, PNET or leptomeningeal
carcinomatosis is required to warrant further investigation. A drug with a mechanism of
action effective at the CSF/leptomeningeal interface would offer a therapeutic advantage.

Methotrexate and Cytosine Arabinoside
Interestingly, the strict application of these criteria led to the exclusion of methotrexate and
cytarabine (in its standard formulation) from the list of potential candidates. We are aware
that these agents are commonly used intrathecally. The application of ideal criteria which
identifies lack of chemosensitivity data and substantial evidence of neurotoxicity led to their
rejection from our ideal list, for the purposes of this review.

Ideal Biological properties
CSF transport system
Some drugs are removed from the CSF by facilitated diffusion using carrier transport systems
such as those for organic ions, quaternary ammonium compounds, organic bases and acids,
also transporters conferring multi-drug resistance such as P-glycoprotein, multi-drug
resistance protein 1, monocarboxylic acid and organic ion transporters [126]. The ideal
intrathecal agent will have no active transport system that is capable of removing it from the

Cell cycle non-specific agent
Cells in the process of dissemination may be in a non-cycling, G
, or slow cycling phase. Drugs
which act at specific stages of the cell cycle will need to be in contact with the cells for
sufficient time for the cells to go through active cell division, to ensure an effective cell kill. An
ideal agent would therefore be cell cycle non-specific in its action.

Ideal Physicochemical and pharmaceutical properties
Active in CSF
Drugs requiring enzyme activation (e.g. cyclophosphamide) are unlikely to be active since
there are very few enzyme systems present in the CSF. Drugs administered in their active
form or with predictable activation (e.g. hydrolysis) are proposed. Those requiring other
processes of activation are excluded.

Factors affecting BBB permeability
Drugs in the CSF are likely to be in equilibrium with plasma as determined by the
characteristics of the BBB and the drugs physical characteristics. Factors that reduce a drugs
capacity to diffuse across the BBB and therefore limit drug efflux include: low lipophilicity, high
hydrophilicity, ionised state at CSF pH of 7.3 and molecular weight >700 Da [53]. Such
properties should therefore enhance sustained CSF drug levels and are therefore preferred for
intrathecal therapy.

Protein binding
Ideally drugs normally protein bound are excluded from crossing the BBB due to molecular
size, direct injection would overcome this and sustain their presence in CSF. This criteria has,
however, not been used as the information was not available for the majority of drugs.

Formulation for CSF administration
The drug must be soluble in the appropriate concentration for intrathecal administration, and
in a vehicle which is suitable for this route, including bio-compatibility with surgical delivery
systems such as ventriculostomy reservoirs.

(Insert Table 1 here - Summary of clinical, biological and physicochemical / pharmaceutical
properties used to justify selection or rejection of drugs)

Literature search strategy and selection criteria
An initial search for the properties of each drug on the candidate list was undertaken by
screening standard pharmaceutical texts [9, 30, 87] and by consulting the manufacturer.
Some information was obtained from these sources. This permitted immediate elimination of a
number of drugs due to evidence of irritancy, neurotoxicity and the need for liver enzyme
activation. However details of ionisation state and degree of protein binding for many drugs
was unavailable despite direct approaches to manufacturers.

For the remaining drugs, Medline (1966 - present) and Embase (1980 present) searches
were performed. Each drug name was separately combined with the following terms:
medulloblastoma, PNET and leptomeningeal carcinomatosis, used in the 'exploded' form.
Where hits exceeded 50 for a combination of terms, the search was further narrowed by
combining that search with intrathecal administration. The full text of papers with relevant
abstracts were obtained and physicochemical information was extracted as well as:
evidence of the drugs activity against medulloblastoma, PNET and leptomeningeal
carcinomatosis in vitro and in vivo
absence or presence of neurotoxicity reported from systemic or intrathecal administration
documentation of intrathecal administration in vivo and in vitro.

One hundred and twenty-six drugs were initially identified for investigation. 35 were
immediately rejected after consulting standard pharmaceutical texts where there was
documentation of irritant qualities, neurotoxicity or the need for enzyme activation of a pro-
drug. Literature searches for the remaining 91 drugs identified 33,627 hits. Screening of
abstracts reduced this to 200 full papers which were read, justifying the rejection of a further
64 drugs (see Table 2a & b).

(Insert Table 2 here List of chemotherapy drugs considered with selection / rejection status
and rejection justification)

Rejected drugs
Of the 126 licensed and investigational anti-cancer chemotherapy agents identified, 99 drugs
were excluded (Table 2a & b). In 45 cases rejection was due solely to the lack of information.
The remaining 54 drugs were rejected for a single reason (n=50) or a combination of two
reasons (n=4). Where a single reason is cited the justifications were: neurotoxicity (n=24),
irritant (n=15), needs activation (n=5), neurotoxic when given intrathecally (n=4), lack of
activity in relevant tumour type (n=2). Where two reasons were cited, the justifications were:
neurotoxic and irritant (n=3), and needs activation and systemic administration results in
equivalent concentration (n=1).

Prioritised drugs for intrathecal use
From our initial list of 126 drugs, 12 drugs have been identified as suitable for further testing
by the intrathecal route in medulloblastoma. Fifteen other potential candidates lack sufficient
information to currently justify exploration of intrathecal administration, but warrant further
research. The 27 drugs are therefore candidates for further testing via the intrathecal route
focussed upon in childhood medulloblastoma and other leptomeningeal malignancies.

The most promising 12 drugs for PNET/medulloblastoma with prior reported clinical experience
are detailed in Table 3 with justification of their potential for future research.

(Insert Table 3 here Drugs currently eligible for consideration for trial by intrathecal
administration for medulloblastoma)

Evaluation in progress
Liposomal cytarabine and mafosfamide are currently undergoing phase 2 trials by the
intrathecal route in children with brain tumours with leptomeningeal spread. Liposomal
cytarabine has now been associated with significant neurotoxicity, particularly in adult patients
also treated with systemic cytarabine [16]. It may be better tolerated in children. Comitant
systemic corticosteroid treatment is recommended. Further studies are needed to establish
optimal use of this drug.

These should be the first drugs to undergo further evaluation in phase 3 studies in order to
further evaluate their efficacy and toxicity profiles and optimise dose and administration

Evaluation required in Phase 2 trials
Carboplatin, etoposide, spartaject, busulfan and nimustine are agents suitable for treatment of
PNET and are therefore also suitable for further investigation by the intrathecal route.
Topotecan is undergoing intrathecal phase 2 studies in children and adults with refractory
neoplastic meningitis; however chemical arachnoiditis has been reported as a dose limiting
toxicity. Etoposide and nimustine have already been tested clinically with some encouraging
results; however refinement of drug delivery systems may be necessary to prolong exposure in
order to overcome their rapid removal from the CSF, their currently observed efficacy may,
therefore, be sub-optimal.

Phase 1 studies required
Floxuridine and 4-hydroperoxycyclophosphamide have been tested intrathecally in limited
studies and have shown promise in relevant tumour types. The intrathecal use of these two
agents should also be further explored.

Pre-clinical evaluation required
Diaziquone is lipophilic and has low ionisation at CSF pH. Together these physical properties
suggest it may not remain in the CSF unless delivered in a sustained release formulation or as
an infusion. It is active in PNET and leptomeningeal carcinomatosis and has been used
intrathecally in human studies. Mercaptopurine has been tested by the intraventricular route
in animals and children with leptomeningeal dissemination of ALL but there is no evidence of
efficacy in primary brain tumours. Rubitecan has been shown to be safe & effective in a rat
model against human glioblastoma multiforme neoplastic meningitis.

Temozolomide is effective in medulloblastoma when given orally. A microcrystalline
preparation has been used intrathecally in animal studies justifying further study if an
appropriate formulation could be developed. Its lipophilicity would favour a sustained release
formulation or an infusion

Pre-clinical animal evaluation
The nine drugs in Table 4 have documented activity in the relevant tumour types but no
information was found on intrathecal use. These drugs would be worthy of intrathecal testing.

(Insert Table 4 here - Drugs with documented activity in the relevant tumour types but no
information on intrathecal use)

The drugs in Table 5 have been tested in studies in the relevant tumour types by the
intrathecal route (some animal, some human) with mixed reports of toxicity and efficacy. With
the exception of bleomycin where neurotoxicity seems to be a major problem, these may well
be worthy of further research. Carmustine is licensed in a sustained release preparation
designed to be implanted into tumour resection cavities as an adjunct to surgery for GBM after
relapse. However, it is documented also to have serious neurotoxicity when given systemically
and therefore has not been prioritised in the list of drugs for further study.

Gemcitabine has been tested by the intrathecal route in a patient with leptomeningeal
carcinomatosis from non small cell lung cancer and has been used intrathecally in non-human
primate models [35, 45, 67]. Based upon this, a phase 1 clinical trial of intrathecal
gemcitabine has been conducted in ten patients with neoplastic meningitis including one with
medulloblastoma. However, dose limiting neurotoxicity was observed in patients who had
previously received CNS-directed therapies [15] The authors concluded that the potential for
severe neurotoxicity precludes further development of gemcitabine for intrathecal

(Insert Table 5 here - Drugs tested by the intrathecal route (some animal, some human) with
mixed reports of toxicity and efficacy)

This work helps draw three important conclusions. First, many of the proposed ideal criteria,
have not been applicable because of inadequate data in the academic and commercial
information domains. A more open shared strategy between pharma and academic
researchers of basic pharmacological data is required. Secondly, using supplementary clinical
criteria, the systematic review has identified 12 drugs suitable for immediate intrathecal use in
medulloblastoma, 15 other candidate drugs were identified for further evaluation prior to
human use. Finally, the extended list of agents, identified here as suitable for intrathecal use,
justifies further efforts to explore their role as part of brain sparing strategies in clinical trials.

Ideal criteria
The pre-determined criteria were applicable in their entirety in only eight drugs resulting in the
current drug selection being based primarily upon alternative criteria including: a) clinical
evidence of tumour chemo-sensitivity, b) previously reported intrathecal use and c) reported
evidence of neurotoxicity after systemic or intrathecal administration. The overwhelming
reason for rejection was because of either insufficient or inconsistent information in the
academic domain (Table 2). It remains unclear why this basic physicochemical data is
unavailable, highlighting the importance of promoting a more open ethic between pharma and
academic groupings exploring novel applications of existing cancer drugs and in particular for
intrathecal administration. This large category of known, rejected drugs as well as a presumed
larger number of unknown candidate agents in industrial archives constitutes an untapped
resource. In contrast, drugs with inconsistent data, but promising early laboratory or clinical
evidence of suitability, are a group where further efforts for their evaluation are justified.
Furthermore, as new biological agents are developed which affect tumour behaviour via non
cytotoxic mechanisms, the lack of publication of physico-chemical criteria or clinical
pharmacological data will slow down the process of their selection or rejection for intrathecal

Drug selection: From this review, 12 agents were identified as potentially suitable for human
use. The three largest previous reviews of this treatment approach do specify optimal drug
criteria for intrathecal use but do not specify the method by which they developed their
proposed drug selections for intrathecal use.

(Insert Table 6 here - Comparison of candidate intrathecal agents by previous published

The current review has identified four additional drugs: carboplatin, floxuridine, 4-
hydroperoxycyclophosphamide and rubitecan, yet rejected cytarabine (in its standard
formulation) and methotrexate because of evidence of neurotoxicity. The extensive experience
with these two latter drugs for prophylactic, therapeutic and palliative strategies in acute
leukaemia and lymphoma has permitted comprehensive reporting of neurotoxicity. It has been
proposed that enhanced and prolonged drug levels due to reduced permeability of the BBB
arising from tumour involvement at sites of CSF efflux is a hypothesis for this observed effect
[23, 24, 55, 86]. Carmustine in its wafer preparation (Gliadel) is used for interstitial therapy
for GBM at relapse (ref). Prior reports of neurotoxicity after intra-arterial administration
might have excluded this drug. Its lipophilicity makes it unsuitable as an intrathecal agent
without adaptation to its formulation to slow its release or as an infusion. Its current
application within a sustained release wafer is, therefore, compatible with the ideal criteria
based upon physicochemical properties but not via the intrathecal route aimed at generalised
CSF distribution.

Nine drugs were identified with evidence of efficacy against medulloblastoma but insufficient
information about their suitability for intrathecal use (Table 4). More information about their
physicochemical / cycle specific and neurotoxicity properties is required before they could be
accepted/rejected for evaluation by the ideal criteria. The information that was missing was
physicochemical (7), cell cycle specificity (4) and neurotoxicity (7). The previous reviews [17,
49, 108] made no suggestions for this category of agents. Drugs, tested by the intrathecal
route but with mixed reports of toxicity (n=7 Table 5), were based upon clinical case reports
from both animal and human experimentation, limiting the scope of this group as a source of
new drugs for testing. There were conflicting reports of neurotoxicity. In only one drug was
there a complete physico-chemical dataset (dacarbazine). In three drugs (bleomycin, busulfan
and carmustine) drug formulation was reported to influence neurotoxicity and efficacy. This
group of drugs highlights the importance of selecting or developing drug formulations that
maximise efficacy and minimise neurotoxicity.

Having identified a short list of 12 drugs suitable for human trial via the intrathecal route
(table 3) an attempt to select those suitable for immediate use has been made based upon
currently available data. Carboplatin, etoposide and nimustine are identified as having the
strongest evidence base for intrathecal use. Liposomal cytarabine (Depocyte) and
mafosfamide are currently undergoing phase 2 trials; their efficacy and toxicity will need to be
judged when given without concomitant chemotherapy or radiotherapy. Mafosfamide is
unlicensed and is therefore unavailable for clinical use in trial or therapy (Personal
communication: Irene Slavc - irene.slavc@akh-wien.ac.at; Dr Susan Blaney:
smblaney@txccc.org). Liposomal cytarabine (Depocyte) is being extensively investigated in adult
studies in a variety of cancer types. Preliminary information has identified substantial
evidence of neurotoxicity [65, 81]. In selecting any of these drugs for use, the more fat
soluble drugs will require either a pharmaceutical formulation aimed at prolonging drug release
or the use of intrathecal infusions to sustain drug levels. Such infusional techniques are well
developed for intrathecal baclofen aimed at controlling spasticity and could be adapted for
such an application [123].

Limitations: Our criteria for accepting or rejecting drugs were applied strictly therefore. Any
drug for which we found documentation of it being irritant or neurotoxic was rejected. It is
possible that this criteria was applied too strictly since, for example, methotrexate and
cytarabine (standard formulation) were excluded and liposomal cytarabine (Depocyte) has
been accepted but evidence of neurotoxicity is now emerging. However the objective of the
study was to identify drugs with ideal characteristics for intrathecal use.

It is difficult to interpret many studies of intrathecal chemotherapy in terms of efficacy, as in
many cases simultaneous systemic chemotherapy or radiation, or both were administered
along with intrathecal treatment making it unclear which component of therapy produced the
effect. Most patients recruited into such studies are refractory to other treatments and are
therefore undergoing such trials as a last resort. Despite this, measurable response to IT
therapy, introduced after systemic chemotherapy has proved refractory, is reported as
evidence of effect.

Drug administration: The clinical benefits of intrathecal administration need to be balanced
against the risks relating to safe prescribing, clinical governance, technical and toxicity
issues [38]. Furthermore, in contrast to leukaemia and lymphomas, primary CNS tumours
represent anatomical challenges to drug distribution linked to the presence of ventrico
peritoneal (VP) shunts, the influence of drainage of the third ventricle to basal cistern by
neuroendoscopic third ventriculostomy, distortion of CNS anatomy linked to post-resection
tumour bed, post operative incarceration of the posterior fossa and spinal blocks. Finally,
the ideal timing of intrathecal administration in relation to debulking surgery is unclear.
Such factors will clearly influence the distribution of CSF and any drug it carries. The
challenges of drug delivery to the brain have resulted in a considerable volume of research,
which may overcome some of these problems, through either local delivery approaches
[105, 140] or vascular delivery across the blood brain barrier [57, 79, 103]. However,
relatively little research has been focussed upon delivery systems for intrathecal delivery,
the main example being that of liposomal cytarabine [22]. The role of efflux inhibitors such
as valspodar in combination with intrathecal therapy in order to overcome the problems of
removal of drugs from the CSF by multi-drug resistance transporters could also be explored

The application of this method of drug administration in medulloblastoma is timely, as cure
rates with combined chemo-radiotherapy for both localised and metastatic cases are improved
(>70%, 5 year survival) in school age children. Novel biological markers are identifying
patients with favourable tumour types where further de-escalation of neuro-toxic cranial
radiotherapy by adjustments to dose/fractionation schedules or fields. The role of systemic
chemotherapy, whilst established in this age group, will be further examined in trials focussing
upon minimising toxicity.

In pre-school age children with favourable presentations (non-metastatic, completely resected,
desmoplastic histology) survival rates also exceed 70% in recently reported studies [54, 109]
using both low and high intensity chemotherapy schedules with or without intrathecal
methotrexate and with or without involved field radiotherapy. The interaction between
biological markers and these treatment strategies in this age group is yet to be studied [8, 58,
110]. The role of chemotherapy is, therefore, established. The potential for novel biological
agents is yet to be explored but is anticipated.

Selecting drugs for IT trial
The most effective and safe way of delivering a drug to the leptomeningeal CSF interface will
continue to be a challenge. In seeking to optimise drug delivery, drug selection is the first
step. The route and method of administration will need to be studied further i.e. the
intraventricular route, the lumbar route, by bolus and by infusion. Anatomical factors will
require consideration including the effects of tumour location, CSF circulation, including VP
shunting and post operative complications affecting the tumour bed. Once drugs have been
tested in phase 1 studies, then phase 2 studies looking for efficacy in patients with either
newly presenting or relapsed leptomeningeal disease will be necessary. Once a drug has been
assessed as effective by intermittent or infusional techniques of IT administration, then phase
3 trials to evaluate its clinical effectiveness or preparation of sustained release formulation for
intermittent IT administration would be justified. The preparatory clinical testing can only be
developed within centres with the relevant neuroscience / pharmaceutical / pharmacological
research expertise.

This review has extended the list of candidate agents suitable for further investigation of
intrathecal treatment of leptomeningeal medulloblastoma. It has identified the importance of
making basic physico-chemical criteria available to the academic domain by pharma and
academic research groupings in order to accelerate the development of new cancer agents. It
has highlighted the opportunity to replicate the success of CNS targeting of drug therapy in
leukaemia and lymphoma, by proposing trials of novel intrathecal chemotherapy in
medulloblastoma. It has identified the need to establish an acceptable process for selecting
new drugs for trial and a suitable trial design.


This work has been supported by funding from the Childrens Brain Tumour Research Centre at
the University of Nottingham.

We would like to acknowledge the assistance of Professor Imti Choonara in the development of
strategy and review of the manuscript, Steve Barrett in the conduct of the literature searching
and Sue Franklin in the preparation of the manuscript.

(Word count 3919)


[1] Interactive analysis LogP and LogW Predictors. www.logp.com.
[2] (1998) Gliadel wafers for treatment of brain tumors. The Medical Letter 40: 92
[3] Aaron R, Elion G, Colvin O, Graham M, Keir S, Bigner D, Friedman H (1994) Busulfan therapy of central
nervous system xenografts in athymic mice. Cancer Chemotherapy & Pharmacology 35: 127-131
[4] Allinson R, Stach P (1978) Intrathecal drug therapy. Drug Intelligence and Clinical Pharmacy 12: 347-359
[5] Archer G, Sampson J, McLendon R, Friedman A, Colvin O, Rose M, Sands H, McCullough W, Fuchs H,
Bigner D, Friedman H (1999) Intrathecal busulfan treatment of human neoplastic meningitis in athymic nude rats.
Journal of Neuro-Oncology 44: 233-241
[6] Arita N, Ushio Y, Hayakawa T, Nagatani M, Huang T-Y, Izumoto S, Mogami H (1988) Intrathecal ACNU - a
new therapeutic approach against malignant leptomeningeal tumors. Journal of Neuro-Oncology 6: 221-226
[7] Arndt C, Colvin O, Balis F, Lester C, Johnson G, Poplack D (1987) Intrathecal administration of 4-
hydroperoxycyclophosphamide in rhesus monkeys. Cancer Research 47: 5932-5934
[8] Ashley D, Merchant T, Zhou T, Coleman L, Pollack I, Duffner P, Burger P, Miller D, Buxton A, Douglas S
(2008) Systemic chemotherapy, second-look surgery and conformal radiation therapy limited to the posterior fossa and
primary site for children >8 months and <3 years with non-metastatic medulloblastoma: A Children's Oncology Group
phase III study. Neuro-Oncology 10: 436-437
[9] Association of the British Pharmaceutical Industry (1999) ABPI Compendium of data sheets and summaries of
product characteristics 1999-2000. Datapharm Publications, London.
[10] Baker S, Heideman RL, Crom WR, Kuttesch J, Gajjar A, Stewart C (1996) Cerebrospinal fluid
pharmacokinetics and penetration of continuous infusion Topotecan in Children with central nervous system tumours.
Cancer Chemotherapy & Pharmacology 37: 195-202
[11] Bates S, Raphaelson M, Price R, McKeever P, Cohen S, Poplack D (1985) Ascending myelopathy after
chemotherapy for central nervous system acute lymphoblastic leukemia: correlation with cerebrospinal fluid myelin
basic protein. Medcial & Pediatric Oncology 13: 4-8
[12] Bayer (1999) Information from Medical Information Department. Newbury, Berkshire.
[13] Bender J, Grillo-Lopez A, Posada Jnr J (1983) Diaziquone (AZQ). Investigational New Drugs 1: 71-84
[14] Berg S, Balis F, Zimm S, Murphy R, Holcenberg J, Sato J, Reaman G, Steinherz P, Gillespie A, Doherty K,
Poplack D (1992) Phase I/II trial and pharmacokinetics of intrathecal diaziquone in refractory meningeal malignancies.
Journal of Clinical Oncology 10: 143-148
[15] Bernardi R, Bomgaars L, Fox E, Balism F, Egorin M, Lagattuta T, Aikin A, Whitcome P, Renbarger J,
Lieberman F, Berg S, Blaney S (2008) Phase I clinical trial of intrathecal gemcitabine in patients with neoplastic
meningitis. Cancer Chemotherapy & Pharmacology 62: 355-361
[16] Bhojwani D, Pui C-H (2008) Intrathecal liposomal cytarabine: more friend than foe? Leukaemia & Lymphoma
49: 1427-1430
[17] Blaney S, Boyett J, Friedman H, Gajjar A, Geyer R, Horowitz M, Hunt D, Kieran M, Kun L, Packer R,
Phillips P, Pollack M, Prados M, Heideman R (2005) Phase I clinical trial of mafosfamide in infants and children aged 3
years or younger with newly diagnosed embryonal tumors: A Pediatric Brain Tumor Consortium study (PBTC-001).
Journal of Clinical Oncology 23: 525-531
[18] Blaney S, Cole D, Godwin K, Sung C, Poplack D, Balis F (1995) Intrathecal administration of topotecan in
nonhuman primates. Cancer Chemotherapy & Pharmacology 36: 121-124
[19] Blaney S, Heideman R, Berg S, Adamson P, Gillespie A, Geyer J, Packer R, Matthay K, Jaeckle K, Cole D,
Kuttesch N, Poplack D, Ballis F (2003) Phase I clinical trial of intrathecal topotecan in patients with neoplastic
meningitis. Journal of Clinical Oncology 21: 143-147
[20] Blaney S, Poplack D (1996) Pharmacologic strategies for the treatment of meningeal malignancy.
Investigational New Drugs 14: 69-85
[21] Blaney S, Poplack D (1998) New cytotoxic drugs for intrathecal administration. Journal of Neuro-Oncology
38: 219-223
[22] Bleyer W (1999) Intrathecal depot cytarabine therapy: a welcome addition to a limited armametarium. Clinical
Cancer Research 5: 3349-3351
[23] Bleyer W, Coccia P, Sather H, et al (1983) Reduction in central nervous system leukemia with a
pharmacokinetically derived intrathecal methotrexate dosage regimen. Journal of Clinical Oncology 1: 317-325
[24] Bleyer W, Drake J, Chabner B (1973) Neurotoxicity and elevated cerebrospinal fluid methotrexate
concentration in meningeal leukemia. New England Journal of Medicine 289: 770-773
[25] Boiardi A, Eoli M, Pozzi A, Salmaggi A, Broggi G, Silvani A (1999) Locally delivered chemotherapy and
repeated surgery can improve survival in glioblastoma patients. Ital J Neurol Sci 20: 43-48
[26] Boland I, Vassal G, Morizet J, Terrier-Lacombe M-J, Valteau-Couanet D, Kalifa C, Hartmann O, Gouyette A
(1999) Busulphan is active against neuroblastoma and medulloblastoma xenografts in athymic mice at clinically
achievable plasma drug concentrations. British Journal of Cancer 79: 787-792
[27] Bomgaars L, Geyer J, Franklin J, Dahl G, Park J, Winick N, Klenke R, Berg S, Blaney S (2004) Phase I trial of
intrathecal liposomal cytarabine in children with neoplastic meningitis. Journal of Clinical Oncology 19: 3916-3921
[28] Brandes A, Pasetto L, Monfardini S (2000) New drugs in recurrent high grade gliomas. Anti-Cancer Research
20: 1913-1920
[29] Brem H, Piantadosi S, Burger P, Walker M, Selker R, Vick N, Black K, Sisti M, Brem S, Mohr G, Muller P,
Morawetz R, Schold S (1995) Placebo-controlled trial of safety and efficacy of intraoperative controlled delivery by
biodegradable polymers of chemotherapy for recurrent gliomas. Lancet 345: 1008-1012
[30] British Medical Association & Royal Pharmaceutical Society of Great Britain (2000) British National
Formulary. BMA & RPS (GB), London.
[31] Castleberry R, Ragab A, Steuber C, Kamen B, Toledano S, Starling K, Norris D, Burger P, Krischer J (1990)
Aziridinylbenzoquinone (AZQ) in the treatment of recurrent pediatric brain and other malignant solid tumors.
Investigational New Drugs 8: 401-406
[32] Chamberlain M, Khatibi S, Kim J, Howell S, Chatelut E, Kim S (1993) Treatment of Leptomeningeal
Metastasis with Intraventricular Administration of Depot Cytarabine (DTC101). Archives of Neurology 50: 261-264
[33] Chamberlain M, Kormanik P, Howell S, Kim S (1995) Pharmacokinetics of Intralumbar DTC-101 for the
Treatment of Leptomeningeal Metastases. Archives of Neurology 52: 912-917
[34] Champagne M, Silver H (1992) Intrathecal Dacarbazine Treatment of Leptomeningeal Malignant Melanoma.
Journal Of The National Cancer Institute 84: 1203-1204
[35] Chen Y-M, Chen M-C, Tsai C-M, Perng R-P (2003) Intrathecal gemcitabine chemotherapy for non-small cell
lung cancer patients with meningeal carcinomatosis - a case report. Lung Cancer 40: 99-101
[36] Crawford S, Rustin G, Bagshawe K (1986) Acute neurological toxicity of intrathecal cytosine arabinoside.
Cancer Chemotherapy & Pharmacology 16: 306-307
[37] Damon L, Plunkett W, Linker C (1991) Plasma and Cerebrospinal Fluid Pharmacokinetics of 1- - D-
Arabinofuranosylcytosine and 1- - D-Arabinofuranosyluracil following the Repeated Intravenous Administration of
High- and Intermediate-Dose - D-Arabinofuranosylcytosine. Cancer Research 51: 4141-4145
[38] Department of Health (2008) Updated national guidance on the safe administration of intrathecal
chemotherapy. In: Health Do (ed).
[39] Dodion P, Sanders C, Georges P, Kenis Y (1988) In vitro chemosensitivity of brain tumors to cisplatin and its
analogues, iproplatin and carboplatin. Cancer Chemotherapy & Pharmacology 22: 80-82
[40] Dollery C (ed) (1999) Therapeutic drugs. Churchill Livinstone, Edinburgh.
[41] Doz F, Pinkerton R (1994) What is the Place of Carboplatin in Paediatric Oncology? European Journal of
Cancer 30A: 194-201
[42] Drayton C (ed) (1990) Comprehensive medicinal chemistry. Pergamon Press, Oxford.
[43] Eagan R, Dinapoli R, Cascino T, Scheithauer B, O'Neill B, O'Fallon R (1987) Comprehensive phase II
evaluation of Aziridinylbenzoquinone (AZQ, Diaziquone) in recurrent human primary brain tumors. Journal of Neuro-
Oncology 5: 309-314
[44] Edwards M, Levin V, Seager M, Wilson C (1981) Intrathecal Chemotherapy for Leptomeningeal
Dissemination of Medulloblastoma. Child`s brain 8: 444-451
[45] Egorin M, Zuhowski E, McCully C, Blaney S, Kerr J, Berg S, Ballis F (2002) Pharmacokinetics of intrathecal
gemcitabine in non-human primates. Clinical Cancer Research 8: 2437-2442
[46] Englehard H (2000) The role of interstitial BCNU chemotherapy in the treatment of malignant glioma. Surg
Neurol 53: 458-464
[47] Ettinger L, Ru N, Krailo M, Ruccione K, Krivit W, Hammond G (1990) A phase II study of diaziquone in
children with recurrent or progressive primary brain tumors: A report from the Children's Cancer Study Group. Journal
of Neuro-Oncology 9: 69-76
[48] Ewend M, Sampath P, Williams J, Tyler B, Brem H (1998) Local delivery of chemotherapy prolongs survival
in experimental brain metastases from breast carcinoma. Neurosurgery 43: 1185-1193
[49] Fleischhack G, Jaehde U, Bode U (2005) Pharmacokinetics following intraventricular administration of
chemotherapy in patients with neoplastic meningitis. Clin Pharmacokinet 4: 1-31
[50] Friedman H (2000) Temozolomide in early stages of newly diagnosed malignant glioma and neoplastic
meningitis. Seminars in Oncology 27: 35-40
[51] Friedman H, Bigner S, Bigner D (1995) Cyclophosphamide therapy of medulloblastoma: from the laboratory
to the clinic and back again (and again and again). Journal of Neuro-Oncology 24: 103-108
[52] Friedman H, Ochs J, Finlay J, Geyer R, Arndt C, Cohen B, Phillips P, Strauss L, Hochberg F, Schold S, Bigner
D, Colvin O (1993) Phase I trial of intrathecal 4-hydroperoxycyclophosphamide for neoplastic meningitis. American
Association for Cancer Research. p. 269.
[53] Green L (1988) Choice of diluents for intrathecal drug administration. Clinical Pharmacy 7: 265-266
[54] Grill J, Sainte-Rose C, Jouvet A, Gentet J-C, Lejars O, Frappaz D, Doz F, Rialland X, Pichon F, Bertozzi A-I,
Chastagner P, Couanet D, Habrand J-L, Raquin M-A, Le Deley M-C, Kalifa C (2005) Treatment of medulloblastoma
with postoperative chemotherapy alone: an SFOP prospective trial in young children. Lancet Oncology 6: 573-580
[55] Grossman S, Trump DL, Chen D, Thompson G, Camargo E (1982) Cerebrospinal fluid flow abnormalities in
patients with neoplastic meningitis. An evaluation using
Indium DTPA ventriculography. American Journal of
Medicine 73: 641-647
[56] Gururangan S, Petros W, Young Poussaint T, Hancock M, Phillips P, Friedman H, Bomgaars L, Blaney S, Kun
L, Boyett J (2006) Phase I trial of intrathecal Spartaject Busulfan in children with neoplastic meningitis: a Pediatric
Brain Tumor Consortium study (PBTC-004). Clinical Cancer Research 12: 1540-1546
[57] Gutman R, Peacock G, Lu D (2000) Targeted drug delivery for brain cancer treatment. J Controlled Release
65: 31-41
[58] Haberler C, Varlet P, Legoix P, Fattet S, Janoueix-Lerosey I, Lellouch-Tubiana A, Grill J, Doz F, Sainte-Rose
C, Delattre O (2008) Widespread nuclear beta-catenin expression in medulloblastoma correlates with mutation status of
CTNNB1 gene encoding beta-catenin. Neuro-Oncology 10: 470
[59] Hare C, Elion G, Houghton P, Houghton J, Keir S, Marcelli S, Bigner D, Friedman H (1997) Therapeutic
efficacy of the topoisomerase I inhibitor 7-ethyl-10-(4-[1-piperidino]-1-piperidino)-carbonyloxy-camptothecin against
pediatric and adult central nervous system tumor xenografts. Cancer Chemotherapy & Pharmacology 39: 187-191
[60] Heideman RL, McCully C, Balis F, Poplack D (1993) Cerebrospinal fluid pharmacokinetics and toxicology of
intraventricular and intrathecal arabinosyl-5-azacytosine (fazarabine, NSC 281272) in the nonhuman primate.
Investigational New Drugs 11: 135-140
[61] Hori T, Tanaka S, Nishiyama M, Kamitani H, Watanabe T, Tabuchi F, Tatsuhara T, Nakajima E (1993)
Distribution of intrathecally administered ACNU in mongrel dogs: Pharmacokinetics and quantitative autoradiographic
study. Surg Neurol 40: 183-195
[62] Houchens D, Ovejera A, Riblet S, Slagel D (1983) Human brain tumor xenografts in nude mice as a
chemotherapy model. European Journal of Cancer & Clinical Oncology 19: 799-805
[63] Huang T-Y, Howng S-L (1997) Intrathecal ACNU against experimental leptomeningeal tumors. Kaohsiung J
Med Sci 13: 626-630
[64] Hussain M, Wozniak A, Edelstein M (1993) Neurotoxicity of antineoplastic agents. Critical Reviews in
Oncology/Hematology 14: 61-75
[65] Jabbour E, O'Brien S, Kantarjian H, Garcia-Manero G, Ferrajoli A, Ravandi F, Cabanillas M, Thomas DA
(2007) Neurologic complications associated with intrathecal liposomal cytarabine given prophylactically in
combination with high-dose methotrexate and cytarabine to patients with acute lymphocytic leukemia. Blood 109:
[66] Jaeckle K, Batchelor T, O'Day S, Phuphanich S, New P, Lesser G, Cohn A, Gilbert M, Aiken R, Heros D,
Rogers L, Wong E, Fulton D, Gutheil J, Baidas S, Kennedy J, Mason W, Moots P, Russell C, Swinnen L, Howell S
(2002) An open label trial of sustained-release cytarabine (DepoCyt) for the intrathecal treatment of solid tumor
neoplastic meningitis. Journal of Neuro-Oncology 57: 231-239
[67] Kerr J, Berg S, Dauser R, Nuchtern J, Egorin M, McGuffey L, Aleksic A, Blaney S (2001) Plasma and
cerebrospinal fluid pharmacokinetics. Cancer Chemotherapy & Pharmacology 47: 411-414
[68] Kim S, Chatelut E, Kim J, Howell S, Cates C, Kormanik P, Chamberlain M (1993) Extended CSF Cytarabine
Exposure Following Intrathecal Administration of DTC 101. Journal of Clinical Oncology 11: 2186-2193
[69] Kim S, Khatibi S, Howell S, McCully C, Balis F, Poplack D (1993) Prolongation of Drug Exposure in
Cerebrospinal Fluid by Encapsulation into DepoFoam. Cancer Research 53: 1596-1598
[70] Kim S, Kim D, Geyer M, Howell S (1987) Multivesicular Liposomes Containing 1- - D-
Arabinofuranosylcytosine for Slow-Release Intrathecal Therapy. Cancer Research 47: 3935-3937
[71] Kimler B, Liu C, Evans R, Morantz R (1992) Intracerebral chemotherapy in the 9L rat brain tumor model.
Journal of Neuro-Oncology 14: 191-200
[72] Kitamura I, Kochi M, Matsumoto Y, Ueoka R, Kuratsu J-I, Ushio Y (1996) Intrathecal chemotherapy iwth 1,3-
Bis(2-chloroethyl)-1-nitrosourea encapsulated into hybrid liposomes for meningeal gliomatosis: An experimental study.
Cancer Research 56: 3986-3992
[73] Kleinschmidt-DeMasters B, Yeh M (1992) "Locked-in syndrome" after intrathecal cytosine arabinoside
therapy for malignant immunoblastic lymphoma. Cancer 70: 2504-2507
[74] Kochi M, Kuratsu J-I, Mihara Y, Takaki S, Inoue N, Sueyoshi N, Uemura S, Ushio Y (1990) Neurotoxicity
and pharmacokinetics of intrathecal perfusion of ACNU in dogs. Cancer Research 50: 3119-3123
[75] Kochi M, Kuratsu J-I, Mihara Y, Takaki S, Seto H, Uemura S, Ushio Y (1993) Ventriculolumbar Perfusion Of
3-[(4-amino-2-methyl-5-pyrimidinyl)methyl]-1-(chloroethyl)-1-nitrosourea hydrochloride. Neurosurgery 33: 817-823
[76] Kochi M, Takaki S, Kuratsu J-I, Seto H, Kitamura I, Ushio Y (1994) Neurotoxicity and pharmacokinetics of
ventriculolumbar perfusion of methyl 6-[3-(2-chloroethyl)-3-nitrosoureido]-6-deoxy-alpha-D-gluco-pyranoside
(MCNU) in dogs. Journal of Neuro-Oncology 19: 239-244
[77] Kohn F, Malkmus S, Brownson E, Rossi S, Yaksh T (1998) Fate of the predominant phospholipid component
of DepoFoam
drug delivery matrix after intrathecal administration of sustained-release encapsulated cytarabine in
rats. Drug Delivery 5: 143-151
[78] Kooistra K, Rodriguez M, Powis G (1989) Toxicity of intrathecally administered cytotoxic drugs and their
antitumor activity against an intrathecal Walker 256 carcinosarcoma model for meningeal carcinomatosis in the rat.
Cancer Research 49: 977-982
[79] Kreuter J (2001) Nanoparticulate systems for brain delivery of drugs. Advanced Drug Delivery Reviews 47:
[80] Kuhl J, Doz F, Taylor R (2004) Embryonic tumors. In: Walker D, Perilongo G, Punt J, Taylor R (eds). Brain &
Spinal Tumors of Childhood. Arnold, London. pp. 314-330.
[81] Lassaletta A, Lopez-Ibor B, Mateos E, Gonzalez-Vicent M, Perez-Martinez A, Sevilla J, Diaz M, Madero L
(2009) Intrathecal liposomal cytarabine in children under 4 years with malignant brian tumours. Journal of Neuro-
Oncology Epub ahead of print
[82] Levin V, Byrd D, Sikic B, Etiz B, Campbell J, Borcich J, Davis R (1985) Central Nervous System Toxicity
and Cerebrospinal Fluid Pharmacokinetics of Intraventricularly Administered Bleomycin In Beagles. Cancer Research
45: 3810-3815
[83] Levin V, Chamberlain M, Silver P, Rodriguez L, Prados M (1989) Phase I/II study of intraventricular and
intrathecal ACNU for leptomeningeal neoplasia. Cancer Chemotherapy & Pharmacology 23: 301-307
[84] Levin V, Edwards M, Gutin P, Vestnys P, Fulton D, Seager M, Wilson C (1984) Phase II evaluation of
dibromodulcitol in the treatment of recurrent medulloblastoma, ependymoma and malignant astrocytoma. Journal of
Neurosurgery 61: 1063-1068
[85] Lind M (1993) Pharmacokinetics of alkylating agents. Pharmacokinetics and Cancer Chemotherapy 17: 157-
[86] MacDonald D (1991) Neurologic complications of chemotherapy. Neurol Clin 9: 955-967
[87] Martindale (1999) The complete drug reference. Pharmaceutical Press, London.
[88] McKeran R, Firth G, Oliver S, Uttley D, O'Laoire S (1985) A potential application for the intracerebral
injection of drugs entrapped within liposomes in the treatment of human cerebral gliomas. Journal of Neurology,
Neurosurgery and Psychiatry 48: 1213-1219
[89] McVie J (1992) Teniposide (VM-26) in brain tumors. Seminars in Oncology 19: 85-88
[90] Morantz R, Kimler B, Vats T, Henderson S (1983) Leomycin and brain tumors. Journal of Neuro-Oncology 1:
[91] Nakagawa H, Maeda N, Tsuzuki T, Suzuki T, Hirayama A, Miyahara E, Wada K (2001) Intracavitary
chemotherapy with 5-fluoro-2'-deoxyuridine (FdUrd) in malignant brain tumors. Japan Journal of Clinical Oncology 31:
[92] Nakagawa H, Miyahara E, Suzuki T, Wada K, Tamura M, Fukushima Y (2005) Continuous intrathecal
administration of 5-fluoro-2'-deoxyuridine for the treatment of neoplastic meningitis. Neurosurgery 57: 266-280
[93] Nakagawa H, Yamada M, Fukushima M, Ikenaka K (1999) Intrathecal 5-fluoro-2'-deoxyuridine (FdUrd) for
the treatment of solid tumor neoplastic meningitis: an in vivo study. Cancer Chemotherapy & Pharmacology 43: 247-
[94] Nakagawa H, Yamada M, Maeda N, Iwatsuki K, Hirayama Z, Ikenaka K (1999) Clinical trial of intrathecal
administration of 5-fluoro-2'-deoxyuridine for treatment of meningeal dissemination of malignant brain tumors. Journal
of Neuro-Oncology 45: 175-183
[95] National Cancer Institute www.nci.nih.gov.
[96] National Library of Medicine Special Information Services http:\\chem.sis.nlm.nih.gov.
[97] Newlands E, Stevens M, Wedge S, Wheelhouse R, Brock C (1997) Temozolomide: a review of its discovery,
chemical properties, pre-clinical development and clinical trials. Cancer Treatment Reviews 23: 35-61
[98] Nishio S, Morioka T, Takeshita I, Tadahisa S, Takanori I, Fujiwara S, Fukui M (1995) Chemotherapy for
progressive pilocytic astrocytomas in the chiasmo-hypothalamic regions. Clinical Neurology and Neurosurgery 87: 300-
[99] Ochiai H, Pernell C, Archer G, Chewning T, McLendon R, Friedman H, Sampson J (2006) Treatment of
neoplastic meningitis with intrathecal 9-Nitro-camptothecin. Neurol Med Chir (Tokyo) 46: 485-490
[100] Olivi A, Dundan L, Corden B, Lenartz D, Brem H (1993) Direct Delivery Of Platinum Based Neoplastics To
The Central Nervous System : A Toxicity And Ultrastructural Study. Cancer Chemotherapy Pharmacology 31: 449-454
[101] Paal C, Erdelyi-Toth V, Pap E, Csaki C, Ferencz T, Schuler D, Borsi J (1994) Pharmacokinetic studies on
Elobromol in children with brain tumors. Anti-Cancer Drugs 5: 539-547
[102] Pardridge W (1996) Brain Drug Delivery and Blood-Brain Barrier Transport. Drug Delivery 3: 99-115
[103] Pardridge W (2002) Drug and gene targeting to the brain with molecular Trojan horses. Nature Reviews Drug
Discovery 1: 131-139
[104] Phillips P, Than T, Cork L, Hilton J, Carson B, Colvin O, Grochow L (1992) Intrathecal 4-
hydroperoxycyclophosphamide: Neurotoxicity, cerebrospinal fluid pharmacokinetics, and antitumor activity in a rabbit
model of VX2 leptomeningeal carcinomatosis. Cancer Research 52: 6168-6174
[105] Read T-A, Thorsen F, Bjerkvig R (2002) Localised delivery of therapeutic agents to CNS malignancies; old
and new approaches. Current Pharmaceutical Technology 3: 257-273
[106] Rhone Poulenc Rorer (1999) Information from Medical Information Department. West Malling, Kent.
[107] Riccardi R, Riccardi A, Di Rocco C, Carelli G, Tartaglia R, Lasorella A, Servidei T, Mastrangelo R (1992)
Cerebrospinal fluid pharmacokinetics of carboplatin in children with brain tumors. Cancer Chemotherapy and
Pharmacology 30: 21-24
[108] Ruggiero A, Conter V, Milani M, Biagi E, Lazzareschi I, Sparano P, Riccardi R (2001) Intrathecal
chemotherapy with antineoplastic agents in children. Paediatric Drugs 3: 237-246
[109] Rutkowski S, Bode U, Deinlein F, Ottensmeier H, Warmuth-Metz M, Soerensen N, Graf N, Emser A, Pietsch
T, Wolff J, Korkmann R, Kuehl J (2005) Treatment of early childhood medulloblastoma by postoperative
chemotherapy alone. New England Journal of Medicine 352: 978-986
[110] Rutkowski S, Von Hoff K, Emser A, Garre ML, Walker D, Grundy R, Dhall G, Finlay J, Grill J (2008)
Prognostic factors and survival of young children with medulloblastoma: An international meta-analysis. Neuro-
Oncology 10: 437
[111] Sampson J, Archer G, Villavicencio A, McLendon R, Friedman A, Bishop W, Bigner D, Friedman H (1999)
Treatment of neoplastic meningitis with intrathecal temozolomide. Clinical Cancer Research 5: 1183-1188
[112] Savaraj N, Feun L, Lu K, Gray K, Wang C, Loo T (1992) Pharmacology of intrathecal VP-16-213 in dogs.
Journal of Neuro-Oncology 13: 211-215
[113] Savas A, Erdem A, Tun K, Kanpolat Y (2000) Fatal toxic effect of bleomycin in brain tissue after intracystic
chemotherapy for a craniopharyngioma: Case report. Neurosurgery 46: 213-217
[114] Schmandt S, Kuhl J (1998) Chemotherapy as prophylaxis and treatment of meningosis in children less than 3
years of age with medulloblastoma. Journal of Neuro-Oncology 38: 187-192
[115] Schold Jr S, Mahaley Jr M, Vick N, Friedman H, Burger P, DeLong E, Albright Jr R, Bullard D, Khandekar D,
Cairncross J, Macdonald D, Falletta J (1987) Phase II diaziquone-based chemotherapy trials in patients with anaplastic
supratentorial astrocytic neoplasms. Journal of Clinical Oncology 5: 464-471
[116] Schuler D, Somlo P, Borsi J, Paraicz E, Revesz T, Kardos G, Koos R (1988) New drug combination for the
treatment of relapsed brain tumors in children. Pediatric Hematology & Oncology 5: 153-156
[117] Schuler D, Somlo P, Koos R, Kalmanchey R, Paraicz E (1992) Treatment of malignant scala posterior brain
tumors in children: the chemotherapy of relapsed medulloblastoma with a dibromdulcitol-containing drug regime and
pharmacokinetic studies of dibromdulcitol in children. Medical & Pediatric Oncology 20: 312-314
[118] Screnci D, McKeage M, Galettis P, Hambley T, Palmer B, Baguley B (2000) Relationships between
hydrophobicity, reactivity, accumulation and peripheral nerve toxicity of a series of platinum drugs. British Journal of
Cancer 82: 966-972
[119] Selby G, Upchurch C, Townsend J, Eyre H (1994) A phase II evaluation of fazarabine in high-grade gliomas: a
Southwest Oncology Group study. Cancer Chemotherapy & Pharmacology 34: 179-180
[120] Slavc I, Schuller E, Czech T, Hainfellner J, Seidl R, Dieckmann K (1998) Intrathecal mafosfamide therapy for
paediatric brain tumors with meningeal dissemination. Journal of Neuro-Oncology 38: 213-218
[121] Slavc I, Schuller E, Falger J, Gunes M, Pillwein K, Czech T, Dietrich W, Rossler K, Dieckmann K, Prayer D,
Hainfellner J (2003) Feasibility of long-tem intraventricular therapy with mafosfamide (n=26) and etoposide (n=11):
Experience in 26 children with disseminated malignant brain tumors. Journal of Neuro-Oncology 64: 239-247
[122] SmithKline Beecham Pharmaceuticals (1999) Information from Medical Information Department. Welwyn
Garden City.
[123] Stempien L, Tsai T (2000) Intrathecal baclofen pump use for spasticity: a clinical survey. American Journal of
Physical Medicine & Rehabilitation 79: 536-541
[124] Stevens M, Newlands E (1993) From triazines and triazines to temozolomide. European Journal of Cancer 29a:
[125] Stiller C, Bleyer W (2004) Epidemiology. In: Walker D, Perilongo G, Punt J, Taylor R (eds). Brain & Spinal
Tumors of Childhood. Arnold, London. pp. 35-49.
[126] Taylor E (2002) The impact of efflux transporters in the brain on the development of drugs for CNS disorders.
Clinical Pharmacokinetics 41: 81-92
[127] Taylor S (1994) New agents in the treatment of primary brain tumors. Journal of Neuro-Oncology 20: 141-153
[128] Tomlinson F, Lihou M, Smith P (1991) Comparison of in vitro activity of epipodophyllotoxins with other
chemotherapeutic agents in human medulloblastomas. British Journal of Cancer 64: 1051-1059
[129] Tsuboi K, Yoshii Y, Hyodo A, Takada K, Nose T (1995) Leukoencephalopathy associated with intra-arterial
ACNU in patients with gliomas. Journal of Neuro-Oncology 23: 223-231
[130] Tuxen M, Hansen S (1994) Neurotoxicity secondary to antineoplastic drugs. Cancer Treatment Reviews 20:
[131] Ushio Y, Kochi M (1998) Intrathecal perfusion chemotherapy with ACNU against subarachnoid dissemination
of glioma in children. Int J Pediat Hem Oncol 5: 55-56
[132] Ushio Y, Kochi M, Kitamura I, Kuratsu J (1998) Ventriculolumbar perfusion of 3-[(4-amino-2-methyl-5-
pyrimidinyl)-methyl]-1-(2-cloroethyl)-1-nitrosourea hydrochloride for subarachnoid dissemination of gliomas. Journal
of Neuro-Oncology 38: 207-212
[133] van der Gaast A, Sonneveld P, Mans D, Splinter T (1992) Intrathecal administration of etoposide in the
treatment of malignant meningitis: feasibility and pharmacokinetic data. Cancer Chemotherapy & Pharmacology 29:
[134] van der Graaf W, Haaxma-Reiche H, Burghouts J, Postmus P (1993) Teniposide for meningeal carcinomatosis
of small cell lung cancer. Lung Cancer 10: 247-249
[135] Vassal G, Boland I, Santos A, Bissery M, Terrier-Lacombe M, Morizet J, Sainte-Rose C, Lellouch-Tubiana A,
Kalifa C, Gouyette A (1997) Potent Therapeutic Activity of Irinotecan (CPT-11) and its Schedule Dependency in
Medulloblastoma Xenografts in Nude Mice. International Journal of Cancer 73: 156-163
[136] Vassal G, Boland I, Terrier-Lacombe M, Watson A, Margison G, Venuat A, Morizet J, Parker F, Lacroix C,
Lellouch-Tubiana A, Pierre-Kahn A, Poullain M, Gouyette A (1998) Activity of Fotemustine in Medulloblastoma and
Malignant Glioma Xenografts in Relation to O
-Alkylguanine-DNA Alkyltransferase and Alkylpurine-DNA N-
Glycosylase Activity. Clinical Cancer Research 4: 463-468
[137] Vassal G, Pondarre C, Boland I, Cappelli C, Santos A, Thomas C, Lucchi E, Imadalou K, Pein F, Morizet J,
Gouyette A (1998) Preclinical development of camptothecin derivatives and clinical trials in pediatric oncology.
Biochimie 80: 271-280
[138] Vassal G, Terrier-Lacombe M-J, Bissery M, Venuat A, Gyergyay F, Benard J, Morizet J, Boland I, Ardouin P,
Bressac-de-Paillerets B, Gouyette A (1996) Therapeutic activity of CPT-11, a DNA-topoisomerase I inhibitor, against
peripheral primitive neuroectodermal tumour and neuroblastoma xenografts. British Journal of Cancer 75: 537-545
[139] Walker D, Wilne S (2005) Treatment of medulloblastoma in young children. Lancet Oncology 6: 541-542
[140] Wang P, Frazier J, Brem H (2002) Local drug delivery to the brain. Adv Drug Del 54: 987-1013
[141] Ward H (1978) Central nervous system tumors of childhood treated with CCNU, vincristine and radiation.
Medical & Pediatric Oncology 4: 315-320
[142] Watterson J, Toogood I, Nieder M, Morse M, Frierdich S, Lee Y, Moertel C, Priest J (1994) Excessive spinal
cord toxicity from intensive central nervous system-directed therapies. Cancer 74: 3034-3041
[143] Weiss R, Issell B (1982) The nitrosoureas: carmustine (BCNU) and lomustine (CCNU). Cancer Treatment
Reviews 9: 313-330
[144] Yamada M, Nakagawa H, Fukushima M, Shimizu K, Hayakawa T, Ikenaka K (1998) In vitro study on
intrathecal use of 5-fluoro-2'-deoxyuridine (FdUrd) for meningeal dissemination of malignant brain tumors. Journal of
Neuro-Oncology 37: 115-121
[145] Yamasaki T, Kikuchi H, Yamashita J, Asato R, Fujita M (1989) Primary spinal intramedullary malignant
melanoma: case report. Neurosurgery 25: 117-121
[146] Yamashima T, Yamashita J, Shoin K (1990) Neurotoxicity of local administration of two nitrosoureas in
malignant gliomas. Neurosurgery 26: 794-800
[147] Yoshida T, Beuls E, Shimizu K, Koulousakis A, Sturm V (1992) Intrathecal chemotherapy with ACNU for
meningeal gliomatosis. British Journal of Cancer 66: 999-1004
[148] Yoshida T, Shimizu K, Koulousakis A, Sturm V (1992) Intrathecal Chemotherapy With ACNU In A
Meningeal Gliomatosis Rat Model. Journal Of Neurosurgery 77: 778-782
[149] Yoshimura J, Nishiyama K, Mori H, Takahashi H, Fujii Y (2008) Intrathecal chemotherapy for refractory
disseminated medulloblastoma. Child's Nervous System 24: 581-585
[150] Zimm S, Collins J, Curt G, O'Neill D, Poplack D (1984) Cerebrospinal fluid pharmacokinetics of
intraventricular and intravenous aziridinylbenzoquinone. Cancer Research 44: 1698-1701
[151] Zimm S, Collins J, Miser J, Chatterji D, Poplack D (1984) Cytosine arabinoside cerebrospinal fluid kinetics.
Clinical Pharmacological Therapeutics 35: 826-830

Table 1: Summary of ideal clinical, biological and physicochemical / pharmaceutical properties
used to justify selection or rejection of drugs.

Property Required
Neurotoxicity low or absent
Evidence of tumour sensitivity

CSF transport system absent
Cell cycle non-specific agent
Physicochemical & pharmaceutical
Active in CSF

Hydrophilic and/or ionised at CSF pH
therefore low membrane permeability (to
minimise diffusion out of CSF)
Molecular Size (700 Da)
Suitable formulation readily available
Table 2: List of chemotherapy drugs considered with selection / rejection status and rejection justification

Table 2a - Rejected chemotherapy agents plus justification

No tumour

No information

Neurotoxic and
Neurotoxic via
IT route
Aclarubicin Carbetimer AG-337
Capecitabine Acivicin Altretamine Doxorubicin
Amsacrine Mitoguazone Amonafide Cyclophosphamide Asparaginase Chlorambucil Melphalan
Azathioprine Amrubicin Mitomycin Carmofur Mitozantrone
Dactinomycin Azacitidine Procarbazine Cisplatin **Fluorouracil
Daunorubicin Bendamustine Trofosfamide Cladribine
Epirubicin Bisantrene *Thiotepa Docetaxel
Idarubicin Caracemide Doxifluridine
Carboquone Fludarabine
Paclitaxel Chlorozotocin
Pirarubicin Chromomycin Gemcitabine
Piroxantrone Clofarabine Hydroxyurea
Plicamycin Crisnatol Ifosfamide
Streptozocin Decitabine Methotrexate
Valrubicin Echinomycin Mitotane
Zorubicin Edatrexate Oxaliplatin
Enloplatin Pentostatin
Enocitabine Sparfosic acid
Estramustine Spiromustine
Homoharringtonine Suramin
Ilmofosine Tegafur
Improsulfan Vinblastine
Lobaplatin Vindesine
Miboplatin Vinorelbine

* Systemic administration delivers concentration to IT route
** Also irritant

Table 2b Selected drugs

1. Carboplatin Selection Table 3
2. Cytarabine Selection Table 3
3. Diaziquone Selection Table 3
4. Etoposide Selection Table 3
5. Floxuridine (FdUrd) Selection Table 3
6. 4-hydroperoxycyclophosphamide Selection Table 3
7. Mafosfamide Selection Table 3
8. Mercaptopurine Selection Table 3
9. Nimustine (ACNU) Selection Table 3
10. Temozolomide Selection Table 3
11. Topotecan Selection Table 3
12. Eflornithine Selection Table 4
13. Mitolactol (Dibromodulcitol) Selection Table 4
14. Didemnin B Selection Table 4
15. Irinotecan Selection Table 4
16. Lomustine (CCNU) Selection Table 4
17. Lonidamine Selection Table 4
18. Menogaril Selection Table 4
19. Tauromustine Selection Table 4
20. Teniposide Selection Table 4
21. Bleomycin Rejection Table 5 neurotoxic when given IT
22. Busulphan Selection Table 5
23. Carmustine (BCNU) Selection Table 5
24. Dacarbazine Selection Table 5
25. Fazarabine Selection Table 5
26. Ranimustine (MCNU) Selection Table 5
27. Zinostatin Selection Table 5

Table 3: Drugs graded by eligibility for trial by intrathecal administration for medulloblastoma

OWPC = octanol/water partition coefficient >0 = lipophilic; <0 = hydrophilic (different values obtained from different references)
pKa = dissociation constant. pKa <5.3 or >9.3 indicate an ionisation of >99% at CSF pH of 7.3
MW = molecular weightIV = intraventricular. IL = intralumbar. IC = intracavity/tumour. IT = intrathecal
RMM - refractory meningeal malignancy, MB = medulloblastoma, LMC = leptomeningeal carcinomatosis, LMM = leptopmeningeal meningitis, NM = neoplastic meningi tis3a
Agents under trial.
DRUG OWPC pKa MW Evidence of
medulloblastoma (MB /
PNET) chemosensitivity
Evidence of IT use. Cell cycle
Comments Refs
-2.46 4.3 243 Evidence of MB sensitivity,
Randomised Clinical Trial in NM
had 13/18 responders with IT
liposomal v 3/17 with free
Licensed in
meningitis in adults.
Phase 1 trial in children
with NM established
maximum tolerated
dose and showed
benefit in 8/14 pts.
Phase 2 trial is in
S phase Neurotoxic in standard
formulation. Liposomal
preparation used for IT is less
toxic and half life in CSF is
prolonged up to 40 times,
though is less in children than
adults. Dexametasone is
required to avoid
[1, 11, 20,
27, 32, 33,
36, 37, 42,
44, 64, 66,
68-70, 73,
77, 78, 87,
95, 96, 130,
Mafosfamide 0.56,-2.11 401 MB, PNET, ependymoma pts,
LMC -rabbit only
IV -non-human
primate, IV rabbit.
Phase 1 study in
patients with RMM, IV+
IL paediatric patients
showed good effect +
minimal toxicity. Phase
1 study has
determined maximum
tolerated dose in
children <3yrs with
newly diagnosed
embryonal tumours
phase 2 trial in
non Cyclophosphamide derivative,
undergoes spontaneous
hydrolysis to active species
[1, 17, 21,
42, 87, 95,
96, 104, 120,
121, 128]

Table 3b: Agents suitable for clinical trial.
DRUG OWPC pKa MW Evidence of
medulloblastoma (MB /
PNET) chemosensitivity
Evidence of IT use. Cell cycle
Comments Refs
Carboplatin -0.46 371 High grade gliomas, MB, PNET,
IT in rats showed this
to be the least
neurotoxic of the
platins. Neurotoxicity
not seen until lethal
non [39, 41,
95, 96,
100, 107,
118, 127]
Etoposide 0.6 9.7 589 MB IV -dogs, humans IV -
metastatic MB,
malignant meningitis.
Used IV in paediatric
patients with good
effect + minimal
Late S, G2 Poor CSF distribution following
IV administration, possibly due
to rapid CSF clearance due to
lipophilicity and/or efflux by P-
glycoprotein and MDR-
associated protein 1.
Possible problem with IL route
due to concerns of drug and
preservatives causing spinal
cord damage.
[1, 40, 42,
87, 95, 96,
112, 114,
121, 126,
128, 133]
Nimustine (ACNU) 0.39
Water +
309 Pilocytic astrocytoma,
meningeal spread of MB, PNET,
glioblastoma, anaplastic
MC - rat, dog, IV in
meningeal spread of
anaplastic glioma, A
phase1/11 study
involving IV/IL/
IT admin in 21 patients
(including children)
with refractory
disseminated MB
showed efficacy in
some patients
non When given IV, needed infusion
- bolus did not give adequate
subarachnoid spread as CSF
half-life very

High priority for future study in
appropriate delivery system
[6, 42, 61,
63, 74-76,
83, 87, 95,
96, 98,
129, 131,
132, 146-

Table 3c: Drugs requiring further investigation before clinical trial.

DRUG OWPC pKa MW Evidence of
medulloblastoma (MB /
PNET) chemosensitivity
Evidence of IT use. Cell cycle
Comments Refs
Floxuridine (FdUrd) -1.16 7.44 246 MB, glioma LMC -animal +
humans, NM
LMC, NM by continuous
IT infusion.
administration may be
useful in small volume
malignant brain
[1, 42, 87,
91-96, 144]
293 Promising phase 1 trial IT in
Promising phase 1 trial
administration in
non [7, 52, 95,

3d: Drugs requiring further investigation, lower priority.
DRUG OWPC pKa MW Evidence of
medulloblastoma (MB /
PNET) chemosensitivity
Evidence of IT use. Cell cycle
Comments Refs
Diaziquone -0.758 low
al pH
364 9L rat brain tumour, LMC,
GBM, AA, paed brain tumours,
RMM, recurrent glioma phase
11 studies, MB xenograft in
Rat, IV-non-human
primate studies,
Human patients
including children with
RMM phase 1/11
studies good
Low aqueous solubility has
caused formulation problems.
Drug shows high rate of
clearance from CSF
[1, 13, 14,
21, 31, 42,
43, 47, 62,
71, 78, 87,
95, 96, 115,
127, 150]
Mercaptopurine 0.01 7.77, 11.17 170 LMD of ALL IV - monkey then
S - phase Cleared from CSF at
0.63ml/min (bulk flow
0.4ml/min) suggesting an
additional mechanism of
elimination - possibly efflux by
MDR-associated protein 1 and
monocarboxylic acid
(25, 59, 60,
91, 105, 107)
Rubitecan 393 safe & effective in rat model
against human GBM xenograft
IT safe & effective in
rat model of GBM NM
Topotecan 0.83, -
6.35, 10.1 458 ependymoma, MB, high grade
glioma xenografts
IV non-human primate.
IV & IL to children &
adults with refractory
neoplastic meningitis.
Phase 2 study
S-phase Chemical arachnoiditis dose
limiting toxicity. Rapid
elimination from CSF. Novel
mechanism of action
[1, 10, 18-
21, 87, 95,
96, 99, 122,

Table 3e: Drug with insufficient information to grade
DRUG OWPC pKa MW Evidence of
medulloblastoma (MB /
PNET) chemosensitivity
Evidence of IT use. Cell cycle
Comments Refs
Temozolomide -0.58, -
194 CNS tumour xenografts, NM
Phase 2 studies in MB,
astrocytoma, glioma, GBM
microcrystalline prep
with increased
solubility - used IT in
rats with NM + malig
glioma sub arachnoid
Spontaneous conversion to
active mitomycin in physiological
conditions. Highly insoluble in
aqueous solution,
microcrystalline form increased
[1, 42, 50,
87, 95-97,
111, 124]

Table 4: Drugs with documented activity in the relevant tumour types but no information on intrathecal use.
DRUG OWPC pKa MW Evidence of tumour
Previous IT

Cell cycle
Comments Refs
Didemnin B 3.173 1112 Some response seen in GBM
and progressive high grade
Natural marine product, antiviral
and antineoplastic activity
[95, 96, 127]
Eflornithine -2.945 182 In combination with
mitoguazone, response seen
in anaplastic astrocytoma and
[95, 96, 127]
Irinotecan 0.03 (pH 1-6)
0.095 (pH9-12).
At pH7-slightly
more lipophilic
but likely to be
more hydrophilic
than lipophilic
677 peripheral PNET,
neuroblastoma xenografts,
MB, glioma, ependymoma
xenografts. GBM,
ependymoma, MB (animal
models), some evidence of
activity in recurrent glioma
S-phase Requires metabolism by
carboxylesterase in liver & tissues
to SN-38 to provide most of its
cytotoxic activity it is not known if
this would occur in the CSF
[28, 59, 87, 95, 96,
106, 135, 138]
Lomustine (CCNU) 2.629 234 MB, glioma non Alkylating and carbamoylating
[10, 42, 62, 85, 87,
95, 96, 114, 141]
Lonidamine 4.407 321 Prolongation of survival +
rate of 1 year survivors (62 v
35%) supratentorial glioma
Inhibits lactate production causing
interference with energy
metabolism of cancer cells
[87, 95, 96, 127]
Menogaril 1.066 542 High grade glioma, phase 1
and 11 trials give conflicting
? non Anthracycline, therefore likely to be
[42, 87, 95, 96,
-0.426 308 MB, PNET, ependymoma -
single agent, moderate effect
long duration of presence in CSF
with a half life of around 24 hours

[39, 42, 84, 87, 95,
96, 101, 116, 117]
Teniposide -0.03(estimated
by calculation
with low
probably higher)
10.13 657 MB in cell culture As etoposide efflux by P-glycoprotein [39, 42, 87, 89, 95,
96, 128, 134]
Tauromustine -0.299 287 AA, glioblastoma, clinical
improvement seen in 19/46
? non (59, 60, 71)

Table 5: Drugs tested by the intrathecal route (some animal, some human) with mixed reports of toxicity and efficacy:
DRUG OWPC pKa MW Evidence of
Previous IT
IV = intra-

Cell cycle
Neurotoxicity Comments Refs
Bleomycin -2.57 1416 Intracerebral
administration in
rat 9L gliosarcoma.
Administered safely
in liposomes in
human cerebral
Given IC in GBM -
safe but efficacy
<25% increase in
survival in LMC rat
phase 1 study
IC - no toxicity.
Rat, beagle,
human (phase 1
IV), depot
Cycling &
M & G2
Vascular necrosis
after IT (beagle),
death after IC
administration in
Antibiotic - binds to
DNA - strand scission
[25, 42, 78,
82, 88, 90,
95, 96, 113]
Busulphan -0.52 246 MB/Glioma/ependy
moma - mice.
Glioblastoma - rat.
Rat Non Confusion +
seizures - high
Irritant. Bifunctional
alkylating agent. Poorly
water soluble however
(Spartaject) been
developed. Given IT in
rats, non-human
primates, adults with
LM disease + phase 1
trial in children with
LMD from brain
[3, 5, 26,
42, 51, 56,
87, 95, 96,
122, 130]
Carmustine (BCNU) 1.53 214 Intravenous
licensed in
brainstem glioma,
MB, astrocytoma,
ependyoma, and
metastatic brain

licensed in high-
grade malignant
glioma and GBM.

IT - in hybrid
liposomes - rat
+ dog
Non Convulsions,
cerebral oedema in
patients with
implants. No
Adverse Drug
Reactions, non-
toxic in rat/dog in
liposomes. Acute
after intracarotid
hy - coma + death
with high dose,
Alkylating and
carbamoylating agent.
DNA alkylation followed
by protein
Causes local venous
[2, 9, 29,
42, 46, 48,
64, 71, 72,
85, 87, 95,
96, 128,
130, 136,
Dacarbazine -0.24 4.42 182 Melanoma -
including primary
3 case reports
human patients
Non Very toxic in rat.
CNS reaction in one
patient case report
Requires demethylation
for activation, though
dacarbazine itself
shows cytotoxic
causes tissue damage
+ severe pain
[9, 12, 34,
42, 78, 85,
87, 95-97,
Fazarabine 496 Lack of activity in
phase 2 trial high
grade glioma. Solid
tumour activity +
Potentially useful
for NM from a
variety of tumours
reservoir) in
Transient changes
in CSF profiles but
no evidence of
neurotoxicity after
IV in monkey
Cytarabine analogue.
Inhibits DNA synthesis
+ methylation.
Nucleoside metabolite,
intracellular activation
needed by
deoxycitidine kinase.
Clearance from CSF
five times higher than
CSF bulk flow rate in
[60, 95, 96,
-1.29 328 pilocytic
Longer half life
than ACNU in
IT admin in 21
children) with
MB showed
efficacy in
Non Mild histological
changes in dog
brain after IV
admin. IC admin
caused marked
brain oedema +
focal necrosis.
Paraplegia + double
incontinence in
some patients
receiving multiple
Nitrosourea [42, 74, 87,
95, 96, 98,
146, 149]
some. IL bolus injections
Zinostatin Glioblastoma,
single case
No CNS toxicity
when given via
carotid artery
Does not permeate
BBB in normal brain.
Half life CSF=50sec
[74, 87, 95,
96, 98, 146]


Table 6: Comparison of candidate intrathecal agents by previous published reviews
Fleischack [49]. Ruggiero [108]. Blaney [17] Conroy

Cytarabine Cytarabine
Liposomal cytarabine
Liposomal cytarabine
Liposomal cytarabine
Liposomal cytarabine
Diaziquone Diaziquone
Etoposide Etoposide
Floxuridine (FdUrd)
Mafosfamide Mafosfamide Mafosfamide Mafosfamide
Mercaptopurine Mercaptopurine Mercaptopurine
Methotrexate Methotrexate

Nimustine (ACNU) Nimustine (ACNU)
Temozolomide Temozolomide
Topotecan Topotecan Topotecan